Efficiency of Segmenting Zones of Interest in Locally Hhomogeneous Scenes

Abstract

We explore the possibilities of segmenting zones of interest on locally homogeneous scenes by some of the known methods. The segmentation quality is quantified in terms of the probability of misclassifying the pixels making up the zone of interest; and the complexity of the zone is quantified in terms of the signal-to-noise ratio. The results of segmentation are presented as a function of the scene complexity and correlation radius. © 2011 Pleiades Publishing, Ltd.